Persons infected with the human immunodeficiency virus type 1 (HIV-1) frequently harbor virus in their oral fluids. Yet, characteristics of saliva-derived virus have not been reported. This pilot investigation will determine and compare the phenotypic syncytium-inducing SI) and genotypic properties of the HIV-1 gp120 V3 domain in saliva and blood using a recently described V3-specific heteroduplex tracking assay (V3-HTA). This plasma-based assay detects non-syncytium-inducing (NSI) viral variants present early in disease and SI variants that evolve with AIDS, based on DNA sequence variability in the V3 domain (major SI determinant). Aim 1: To test the hypothesis that HIV-1 in the body fluids are genetically related and derived from a common source, paired whole saliva and plasma samples will be collected from 50 infected adults and analyzed by V3 HTA. Briefly, viral RNA is isolated from samples, reverse transcribed (RT) and the V3 domain amplified by PCR. RT-PCR products are denatured and then annealed to a radiolabeled NSI V3 consensus probe. The resulting heteroduplexes are separated by 12 percent PAGE. SI phenotype of heteroduplexes are tentatively assigned based on their V3-HTA migration patterns. Aim 2: To confirm phenotype, DNA and corresponding amino acid sequences of V3 products will be determined and classified as NSI or SI based on amino acid substitution pattern. Aim 3: As a second approach to verify phenotype, recombinant viruses containing twelve (six each of NSI and SI) representative saliva-derived V3 sequences will be generated by site-directed mutagenesis of the HIV-1 NFN-SM clone (NSI phenotype). Viral stocks will be prepared and tested for syncytium formation in the MT-2 T cell assay. Results obtained from the three methods will be compared in statistical analyses. Aim 4: To test the hypothesis that AIDS patients harbor salivary SI variants, data describing disease stage, CD4 cell count, antiretroviral therapy and viral loads in saliva and plasma will be obtained by medical chart review and HIV-1 RNA assay. Data will be correlated with salivary SI phenotype. This study will give insight into the source of HIV-1 in saliva. V3-HTA analysis of saliva may represent a novel, non-invasive method for early detection of impending SI stages in NSI-infected individuals, possibly leading to improved therapies for HIV.